Hypergraph clustering for analyzing chronic disease patterns in mild cognitive impairment reversion and progression.

Area of Science:

• Neurology
• Gerontology
• Data Science

Background:

• Understanding the progression and potential reversal of mild cognitive impairment (MCI) requires identifying preceding chronic condition patterns.
• The Mayo Clinic Study of Aging (MCSA) cohort provides valuable data for analyzing cognitive trajectories and associated health conditions.

Purpose of the Study:

• To identify common sequences of chronic conditions preceding MCI.
• To introduce a novel network-based clustering framework (hypergraph clustering) for characterizing patients with similar progression patterns linked to cognitive trajectories.
• To explore sex-specific indicators associated with MCI reversal.

Main Methods:

• Utilized the Mayo Clinic Study of Aging (MCSA) cohort.
• Categorized participants into groups based on MCI progression or reversion to normal within five years.
• Curated state transition networks at the patient level and applied hypergraph clustering at the patient group level.

Main Results:

• Identified generic key indicators, such as chronic kidney disease, associated with MCI.
• Highlighted sex-specific potential indicators for MCI reversal, including arthritis and hypertension in females.
• Observed specific chronic condition sequences, like those originating from arthritis, more commonly in females, warranting further validation.

Conclusions:

• The novel hypergraph clustering framework effectively characterizes patients by uncovering similarities in unique chronic condition trajectories preceding MCI.
• Findings open new research avenues into sex-specific differences in chronic condition patterns and their impact on MCI progression and reversal.